Despite the success of highly active antiretroviral therapy (HAART) on the population level, as many as 30-70% of patients in the clinical setting fail to achieve sustained virologic responses to HAART and toxicities are common. Treatment outcomes of HAART have many determinants, including the amount of drug that reaches the site of antiviral activity (exposure). Our ability to accurately measure exposure to antiretrovirals (ARVs) is currently limited, most notably secondary to the significant interindividual pharmacokinetic (PK) variability observed for these agents. Traditional methods of therapeutic drug monitoring employ single untimed blood levels of ARVs to assess exposure, which can't account for this interindividual PK variability; hence, these methods have had inconsistent success in predicting treatment responses and have not become routine measures in the clinical setting. Full PK studies for ARVs to date have been generally performed in highly selected patients (typically male) in order to minimize factors that may contribute to PK variability, thus limiting the generalizability of these results to typical patients in clinical care. I propose to employ a relatively novel methodology - population PK modeling - to identify sources of interindividual and intraindividual PK variability for ARVs and to calculate accurate ARV exposure measurements for diverse populations. In a prospective cohort of HIV-infected patients, I propose to quantify the factors that produce interindividual variability in ARV exposure using population PK methods with intensive PK sampling data obtained from a representative sample of HIV-infected women (Aim 1); to test whether ARV exposure estimates derived from sparse sampling, combined with individual covariate data, population models, and appropriate statistical methodologies, predict treatment outcomes more accurately than single plasma ARV levels (Aim 2), and to determine whether biologic sex influences PK variability for ARVs (Aim 3). My mentoring committee consists of internationally recognized scientists whose expertise spans pharmacology (Dr. Benet), epidemiology and observational research (Dr. Greenblatt), and biostatistics (Dr. Bacchetti). Their mentorship, as well as a focused training and research plan facilitated by a K23 award, will enable me to become an independent clinical investigator, focusing on the association of antiretroviral treatment exposure with responses to therapy in diverse populations.